Re: Split off ActivityPub CG or WG

út 3. 10. 2023 v 15:35 odesílatel Evan Prodromou <evan@prodromou.name>
napsal:

> Here's the thing: that CG is not for ActivityPub as we know it!
>
> I started the "ActivityPub" CG before the Social CG and WG. It was made to
> standardise a similar but distinct API, which used the old Activity Streams
> extension to Atom, plus the AtomPub API.
>
> We used it for our native API in StatusNet. I'm not sure it survived the
> name change to GNU Social.
>
> When we standardised ActivityPub at the Social WG, we gave it the same
> name as this obscure API. It was never part of that CG, which never really
> got off the ground.
>
> I'm not sure when the ActivityPub CG was closed; I requested it several
> times.
>

Thank you for the context, Evan. Given the existing support and interest in
ActivityPub, it's logical and straightforward to modernize and refocus the
group. A simple post-activation blurb update ensures alignment with current
purposes. Overlaps in platforms such as CG, GitHub, Matrix, SocialHub, and
the FEP system have historically diversified rather than diluted fediverse
momentum. With substantial support, including from a spec editor, this
approach seems an intuitive and efficient means to consolidate collective
expertise for ActivityPub’s continued development.


>
> Evan
>
> On Oct 3, 2023 03:01, aschrijver <facilitator@humanetech.community> wrote:
>
> +1
>
> As Melvin pointed out an ActivityPub CG already exists, and needs only be
> reactivated. What we see in the enormous amount of discussions in all the
> various channels, is how many challenges and improvements still need to be
> tackled just in the AP protocol space. Something a dedicated CG can way
> more efficiently focus on.
>
> At the same time there's the SocialCG, with a focus on the Social Web as a
> whole. I do not think having an ActivityPub CG obviates the need for a
> SocialCG. Instead I feel that the SocialCG itself may be more efficient if
> it focuses on the higher level concern of bringing different protocols
> closer together.
>
>
> ------- Original Message -------
> On Sunday, October 1st, 2023 at 20:42, Christine Lemmer-Webber <
> cwebber@dustycloud.org> wrote:
>
>
> >
> >
> > I'm not getting too involved. So, you can ignore me. I'm here in the
> > background, peering vaguely, in the few moments I'm on top of my email
> > (not often these days). But, if you want my advice... and maybe you
> > don't...
> >
> > I think it's time for ActivityPub to break off into its own CG or WG.
> > The SocialCG or WG, whatever happens, can be a thing that exists, and
> > ActivityPub people can be part of it, but my experience with the
> > SocialWG especially was that a lack of core agreement on what we were
> > working on really made life incredibly difficult. We got some good work
> > done, but... there's enough to do without needing to have the
> > disagreements that come from not agreeing on fundamentals.
> >
> > I think if a re-invigorated set of ActivityPub work is to happen, do it
> > in a new group devoted to that explicit purpose. You'll retain a lot
> > more hair of everyone participating.
> >
> > Now... regarding the CG or WG process... well, it's been nice seeing
> > just how well WebAssembly is doing with their CG process. That's given
> > me hope. So I think Ben's suggestion is not bad. That said the
> > SocialWG worked pretty well BECAUSE it was full of invited experts. But
> > that was heavily frowned upon by the W3C at the time. If a WG were to
> > happen, get buy-in to that idea up front.
> >
> > But yeah. ActivityPub CG/WG. Keep it focused. Let people get the hard
> > work done they need to when already agreeing on a core basis. Otherwise
> > else it's gonna be just like last time. And that took a few years off
> > my lifespan.
> >
> > Just my opinions,
> > - Christine
>
>
>